cost

Parse local JSONL transcript logs to estimate pay-per-token API costs in USD and EUR.

38|10|Updated Oct 29, 2025
One-click install
npx skills add https://github.com/maxvaega/awesome-skills --skill cost-maxvaega
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: cost
Source: https://github.com/maxvaega/awesome-skills/tree/main/cost
Command: npx skills add https://github.com/maxvaega/awesome-skills --skill cost-maxvaega

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill solves the uncertainty regarding how much your Claude Cowork or Claude Code usage would cost if billed on a pay-per-token basis, providing a transparent comparison against your subscription.

Core Features & Use Cases

  • Faithful Cost Replay: Calculates exact costs based on actual JSONL transcript logs, including prompt-cache hits and token usage.
  • Granular Reporting: Provides token counts and costs in both USD and EUR for individual conversations or your entire history.
  • Use Case: Use this to audit your agent's efficiency or to decide if your current usage patterns are more cost-effective under a subscription or a pay-per-token API model.

Quick Start

Ask the agent to calculate the API cost of the current conversation by typing /cost.

Frequently Asked Questions about cost

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I estimate pay-per-token API costs for Claude Code sessions?

You can estimate pay-per-token API costs for Claude Code sessions by parsing local JSONL transcript logs to analyze input, cache-read, cache-write, and output token usage in USD and EUR.

What is the best way to audit Claude agent token usage and financial reporting?

Auditing Claude agent token usage involves replaying exact costs from JSONL transcript logs, analyzing token counts across caching categories, and comparing the results against your current subscription model.

Does API cost estimation work with historical Claude Cowork conversation logs?

Yes, API cost estimation works with historical Claude Cowork logs by accessing the local ~/.claude/projects directory to provide session-specific or complete historical financial reporting.

How to calculate Claude API costs in USD and EUR from local transcript files?

To calculate Claude API costs in USD and EUR, the agent parses local JSONL transcript files to extract token usage data and applies pay-per-token billing rates for accurate financial reporting.

Can I use transcript logs to decide if a pay-per-token API model is cheaper than a subscription?

Yes, you can use transcript logs to calculate exact pay-per-token API costs based on actual token usage, providing a transparent financial comparison to decide against your subscription.

Do I need access to the ~/.claude/projects directory to analyze Claude API costs?

Yes, you need access to the local ~/.claude/projects directory because the cost analysis requires parsing the JSONL transcript logs stored there to extract accurate token usage data.